A registration method for augmented reality based on template matching

摘要

In this paper, we propose a registration method for augmented reality based on template matching. Some kinds of fiducials such as square or circle have been used by many augmented reality applications. Our new method uses a black square fiducial to obtain the initial condition for tracking, but dose not use it in iterative tracking phase. Several feature points are detected from the tracked object by off-line image analysis. While tracking the object, some of the feature points are selected for template matching and the object pose and position are calculated from the results of feature tracking.